I am trying to create a simulated SNP dataset, consisting of ~40 individuals from two populations, and data from around 1000 co-dominant SNP markers, exhibiting a predefined (low) level of structure. The result I am looking for is one similar to that given by genalex's create function, but the crucial difference is that I require a dataset comprising 4 SNP alleles overall (A,C,G,T), but only two alleles per SNP marker (A/T, C/G, C/T etc.). All combinations present, basically like you would expect from a typical population genetic SNP dataset.
Essentially, I guess I am after a function that allows me to dictate parameters such as level of structure, no. of individuals, number of alleles overall and per SNP, and that delivers a SNP dataset that roughly adheres to those parameters. An R solution would be preferable as it is the only language I am comfortable with for the moment.